Re: pgcc

Marc Lehmann (marc@gimp.org)
Sat, 11 Sep 1999 03:44:27 +0200

On Fri, Sep 10, 1999 at 03:39:08PM +0000, Mark Hahn <hahn@coffee.psychology.mcmaster.ca> wrote:
> > K5, K6, Cyrix and may be for IDT chips. Here the difference can be
> > significant.
>
> please show numbers. I've tried chip-specific compile options before
> and never seen any measurable difference in kernel performance. this
> is exactly what I expected, of course, since the kernel is basically
> not ever compute-bound. do you have specific cases where compilation
> makes a measurable difference in kernel performance?

In my personal opinion I'd say compiling the kernel with pgcc is rarely if
ever going to make it faster (for the reasons you state).

However I would be interested in comparisons for the kernel on the k6,
specially between a pgcc snapshot and pgcc-2.95.1.

pgcc-2.95.1 uses the k6 support that comes with gcc, the original amdk6
support in pgcc was thrown away.

Unfortunately, people claimed that this slowed everything down (with
-mk6), and it would be interesting to be able to measure this on the
kernel (maybe using lmbench).

-- 
      -----==-                                             |
      ----==-- _                                           |
      ---==---(_)__  __ ____  __       Marc Lehmann      +--
      --==---/ / _ \/ // /\ \/ /       pcg@goof.com      |e|
      -=====/_/_//_/\_,_/ /_/\_\       XX11-RIPE         --+
    The choice of a GNU generation                       |
                                                         |


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/